day0-release

Deterministic end-to-end driver for day-0 quantized-checkpoint releases — chains PTQ → evaluation → comparison with enforced gates between stages (the evaluation stage deploys the checkpoint itself), and returns a publish decision (ACCEPT / REGRESSION / ANOMALOUS / INFEASIBLE). Use when the user asks to "release a model at day-0", "quantize and validate model X is within N% of baseline and tell me if it's publishable", or "run the full day-0 workflow". Do NOT use for single-stage requests — quantizing only (use ptq), serving only (use deployment), evaluating only (use evaluation), or comparing two existing runs (use compare-results).

Quality

100%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

100%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is an exceptionally well-structured conductor skill: a fixed gated chain, executable gate commands, an explicit progress checklist, and a failure_class→action triage table, all without padding. It cleanly separates orchestration from the domain skills it delegates to.

Dimension	Reasoning	Score
Conciseness	Lean and information-dense with no padding or explanation of concepts Claude already knows; the chain diagram, step checklist, gate invocations, and triage table each earn their tokens, matching the score-3 anchor.	3 / 3
Actionability	Provides fully executable, copy-paste-ready commands with real flags (e.g. gate_ptq.py --summary, gate_run.py --run, gate_compare.py --baseline/--candidate/--threshold plus the --scales override) and exact decision semantics, per the score-3 anchor.	3 / 3
Workflow Clarity	A clear setup→PTQ→baseline-eval→quantized-eval→compare→closeout sequence with a gate after every stage, an explicit checklist ("Do not advance past a failed gate"), and a Triage table mapping each failure_class to a recovery action — explicit validation checkpoints and feedback loops per the score-3 anchor.	3 / 3
Progressive Disclosure	The body is a conductor overview pointing one level deep to domain skills and references (ptq/references/checkpoint-validation.md, skills/common/*), bundling the three real gate scripts it owns in ./scripts while deferring deeper mechanics elsewhere, matching the clear-overview score-3 anchor.	3 / 3
	Total	12 / 12 Passed

Description

100%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is concise, third-person, and clearly states both what the skill does and when to use it with realistic trigger phrasing and explicit anti-trigger routing. It is among the strongest examples of a multi-stage conductor skill description.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "chains PTQ → evaluation → comparison", "enforced gates between stages", and a concrete "publish decision (ACCEPT / REGRESSION / ANOMALOUS / INFEASIBLE)" — matching the score-3 anchor of multiple specific concrete actions.	3 / 3
Completeness	Explicitly answers what the skill does (deterministic end-to-end driver chaining PTQ→eval→comparison with gates, returning a publish decision) and when to use it via a "Use when…" clause with quoted triggers, satisfying both halves of the score-3 anchor.	3 / 3
Trigger Term Quality	Quotes natural user phrases ("release a model at day-0", "quantize and validate model X is within N% of baseline and tell me if it's publishable", "run the full day-0 workflow") giving good coverage of terms users would actually say, per the score-3 anchor.	3 / 3
Distinctiveness Conflict Risk	A clear day-0 release niche with an explicit "Do NOT use for single-stage requests" routing clause distinguishing it from ptq/deployment/evaluation/compare-results, making it unlikely to trigger for the wrong skill per the score-3 anchor.	3 / 3
	Total	12 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: NVIDIA/Model-Optimizer
Path: .agents/skills/day0-release/SKILL.md
Commit: 87c9f8c

Reviewed: 15 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.